Management Science research is done on three levels:

i) The fundamental level: The foundations of Management Science lie in three mathematical disciplines: Probability, Optimisation, and Dynamical Systems. These disciplines provide us with the language of our field. We use this language, in conjunction with computer science and statistics tools, to build and analyse models of complex management systems. Fundamental research in Management Science will typically improve relevant knowledge of the cognate mathematical disciplines and is often published in applied mathematics journals. An excellent journal for this type of work is Mathematics of Operations Research.

ii) The modelling level: Modelling is at the heart of Management Science. Building models, analysing them mathematically, gathering and analysing data, implementing models on computers, solving them analytically or numerically, playing with them, interpreting the results - all this is part of Management Science research on the modelling level. Cognate disciplines are economics, computer science, numerical analysis and statistics. The flagship journals for this work are Management Science  and Operations Research.

iii) The application level: Management Science, just as any other engineering discipline, has strong aspirations to make a practical impact and be a driver for change in the real world. When you work with a model, there is always a strong temptation to shut the door to reality and focus on the much cleaner model world. A key difference between Management Science and mathematics is that we don't take our eyes off the application. Almost all journals in the field appreciate or even require a credible link to real applications. A good journal that is dedicated to reports on successful Management Science applications is Interfaces.

Students on the MPhil in Management Science typically focus on one of the above levels in their dissertation work. We encourage MPhil students to work in teams, with different team members focusing on the different levels for the same problem area.

Fundamental work. On the fundamental level, we encourage research in optimisation theory and dynamical systems. Students who are primarily interested in fundamental work in probability might wish to explore the MPhil in Statistical Science.

Modelling work. Work on the modelling level includes hierarchical optimisation and game theory, integer programming for scheduling, stochastic dynamic optimisation with application in real options and revenue management, machine learning and data mining, systems dynamics, and demand forecasting. We encourage students to gain a wide knowledge of different types of models and be able to explore and combine a variety of modelling approaches in their research work.

Applications. Most of our modelling work goes hand-in-hand with and is driven by practical applications. We are open to exploring applications across a wide spectrum of functional areas and industries. Examples of current application areas are

- Energy industry (gaming models for bidding behaviour in electricity generation, electricity markets design and regulation, climate change modelling, portfolio management in the oil and gas industry)

- Transportation and aviation (airline revenue management, competition and options products in the airline industry, road congestion charging)

- Health care and life sciences (patient flow modelling, performance analysis and improvements of hospitals, effectiveness of contagious disease management, valuation of drug discovery projects, R&D portfolio management and contract design in the life sciences, bio-informatics, etc.)

- Marketing modelling (Media campaign optimization, demand forecasting for new products, effects of promotions)

- Real portfolio optimization (R&D portfolio management, venture capital portfolio optimization, Exploration portfolio optimization in the petroleum industry)

Strategic industry partners. Our Management Science research agenda, from the most fundamental to the applications level, is inspired and driven by important practical problems. To enforce this point, we maintain strategic partnerships with several large corporates in our core application areas, such as British Airways and Shell, Lloyds TSB, as well as with local small and medium size enterprises, such as the Cambridge University Hospital, and Cambridge Antibody Technology. Discussions with senior managers of these companies are important in informing our research planning.

Management Science meets Business Strategy. At present, we are adding a complementary area to our Management Science work: The use of computer models for strategic management. How can we use models to help a board of directors make a confident choice between strategic options and make good long-term decisions in a highly uncertain, highly dynamic, and highly complex business world? Some key questions are: How can we best communicate modelling work, or the results thereof, so that they are understandable at the board level and to a wider, non-technical audience of stakeholders? What is the added value of an additional layer of technical complexity in our models and how can we trade off the improved relevance of a more complex model against the implied loss of audience? How can we incorporate variables that are less amenable to quantifications without loosing the discipline imposed by a quantitative modelling environment?